# Claude Sonnet 5.5 vs GPT-5.4

> Claude Sonnet 5.5 is the stronger model overall, scoring 61.9 to 59.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-sonnet-5-5-vs-gpt-5-4
- Last updated: 2026-10-12
- Shared benchmarks: 28

## Summary

- They share 28 benchmarks with published results for both. Claude Sonnet 5.5 scores higher in 5 categories and GPT-5.4 in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Claude Sonnet 5.5 leads 67.3 to 52.6.
- The biggest single-benchmark swing is ProofBench: 100% for Claude Sonnet 5.5 and 56% for GPT-5.4.
- Claude Sonnet 5.5 is cheaper at $2 / $10 per million input/output tokens, against $2.50 / $15 for GPT-5.4.
- GPT-5.4 accepts more context: 1.05M tokens versus 1M.

## Snapshot

| | Claude Sonnet 5.5 | GPT-5.4 |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 61.9 | 59.4 |
| Rank | 10 | 16 |
| Context | 1M | 1.05M |
| Input $/M | $2 | $2.50 |
| Output $/M | $10 | $15 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Sonnet 5.5: 67.3 (#6)
- GPT-5.4: 52.6 (#33)

| Benchmark | Claude Sonnet 5.5 | GPT-5.4 |
|---|---|---|
| LMArena WebDev | 1774 | 1465 |
| SciCode | 61% | 56.6% |
| LMArena Coding | 1513 | 1497 |
| ALE-Bench | 1,819 | 1,607 |
| SWE-bench Verified | — | 76.9% |
| DeepSWE | — | 51.8% |
| FrontierCode | 52.1% | — |
| CursorBench | 55.5% | — |
| FrontierSWE | 61.9% | — |
| GSO | — | 31.4% |
| WeirdML | — | 77.7% |
| MirrorCode | — | 15.6% |
| AlgoTune | — | 1.85 |

## Agentic & Tool Use

- Claude Sonnet 5.5: 45.0 (#16)
- GPT-5.4: 46.5 (#13)

| Benchmark | Claude Sonnet 5.5 | GPT-5.4 |
|---|---|---|
| APEX-Agents | 75.5% | 52.4% |
| Terminal-Bench | — | 81.8% |
| τ²-bench Banking | — | 39.4% |
| DeepResearch Bench | — | 35.1% |
| PostTrainBench | — | 19% |
| GBAEval | — | 45.1% |
| LMArena Search | — | 1197 |
| METR Time Horizons | — | 74.3% |
| Vending-Bench 2 | — | 6,144 |

## Reasoning

- Claude Sonnet 5.5: 54.0 (#28)
- GPT-5.4: 61.8 (#19)

| Benchmark | Claude Sonnet 5.5 | GPT-5.4 |
|---|---|---|
| NYT Connections (extended) | 80.5% | 91.3% |
| CritPt | 31.4% | 23.4% |
| LMArena Hard Prompts | 1495 | 1485 |
| Mystery Game Puzzles | 65% | 37% |
| Epoch Capabilities Index | 165.03 | 156.81 |
| ARC-AGI-2 | — | 74% |
| Kagi LLM Benchmark | — | 63.8% |
| ARC-AGI-1 | — | 93.7% |
| Chess Puzzles | — | 44% |
| EnigmaEval | — | 16% |
| Thematic Generalization | — | 80% |
| EBR-Bench | — | 25.4% |
| DTBench | — | 94.4% |
| LMCA | — | 52% |
| ForecastBench | — | 59.5 |

## Math

- Claude Sonnet 5.5: 87.9 (#6)
- GPT-5.4: 73.5 (#19)

| Benchmark | Claude Sonnet 5.5 | GPT-5.4 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 88.8% | 78.6% |
| FrontierMath Tier 4 | 80.5% | 49% |
| OTIS Mock AIME 2024-2025 | 100% | 97.8% |
| ProofBench | 100% | 56% |
| LMArena Math | 1510 | 1488 |
| MathArena Final-Answer Competitions | — | 83.1% |
| FrontierMath (Feb 2025 set) | — | 47.6% |
| FrontierMath Erdős | 2.9% | — |
| FrontierMath Tier 4 (v1) | — | 27.1% |

## Knowledge

- Claude Sonnet 5.5: 66.0 (#12)
- GPT-5.4: 65.3 (#14)

| Benchmark | Claude Sonnet 5.5 | GPT-5.4 |
|---|---|---|
| GPQA Diamond | 95.6% | 93.3% |
| SimpleQA Verified | 46.5% | 45.1% |
| LMArena Expert | 1540 | 1507 |
| Humanity's Last Exam | — | 36.2% |
| Vectara Hallucination Rate | — | 7% |

## Multimodal

- Claude Sonnet 5.5: 51.5 (#6)
- GPT-5.4: 43.7 (#20)

| Benchmark | Claude Sonnet 5.5 | GPT-5.4 |
|---|---|---|
| LMArena Vision | 1289 | 1303 |
| Furniture Assembly | 75% | 37.5% |
| Blueprint-Bench 2 | — | 27.1% |
| LMArena Document | — | 1471 |

## Multilingual

- Claude Sonnet 5.5: 55.3 (#30)
- GPT-5.4: 56.2 (#23)

| Benchmark | Claude Sonnet 5.5 | GPT-5.4 |
|---|---|---|
| LMArena Non-English | 1452 | 1465 |
| LMArena Chinese | 1522 | 1519 |
| LMArena Russian | 1451 | 1480 |
| LMArena French | — | 1493 |
| LMArena German | — | 1472 |
| LMArena Japanese | — | 1485 |
| LMArena Korean | — | 1448 |
| LMArena Spanish | — | 1454 |

## Instruction Following

- Claude Sonnet 5.5: 78.3 (#11)
- GPT-5.4: 77.1 (#27)

| Benchmark | Claude Sonnet 5.5 | GPT-5.4 |
|---|---|---|
| LMArena Instruction Following | 1495 | 1469 |

## Long Context

- Claude Sonnet 5.5: 45.9 (#28)
- GPT-5.4: 50.3 (#8)

| Benchmark | Claude Sonnet 5.5 | GPT-5.4 |
|---|---|---|
| LMArena Longer Query | 1498 | 1473 |
| CL-bench | — | 27.9% |
| CL-bench Life | — | 21.7% |

## Writing & Preference

- Claude Sonnet 5.5: 66.0 (#40)
- GPT-5.4: 71.9 (#17)

| Benchmark | Claude Sonnet 5.5 | GPT-5.4 |
|---|---|---|
| LMArena Text | 1471 | 1469 |
| LMArena Creative Writing | 1465 | 1439 |
| LMArena Multi-Turn | 1474 | 1482 |
| EQ-Bench Creative Writing | — | 1840 |
| EQ-Bench 4 | — | 1272 |

## FAQ

### Is Claude Sonnet 5.5 better than GPT-5.4?

Claude Sonnet 5.5 is the stronger model overall, scoring 61.9 to 59.4 on the Noometry Index.

### Which is cheaper, Claude Sonnet 5.5 or GPT-5.4?

Claude Sonnet 5.5 is cheaper. It lists at $2 per million input tokens and $10 per million output tokens; GPT-5.4 lists at $2.50 and $15.

### Is Claude Sonnet 5.5 or GPT-5.4 better for coding?

Claude Sonnet 5.5 scores higher on coding benchmarks: 67.3 versus 52.6 in the Noometry coding category.

### Which has the bigger context window?

GPT-5.4 does, with 1.05M tokens against 1M.

### How many benchmarks do Claude Sonnet 5.5 and GPT-5.4 share?

28 benchmarks have published results for both models. Claude Sonnet 5.5 has 32 scored results on Noometry and GPT-5.4 has 68.